The Importance of Genes in Inheritance and Phenotype Determination
A century of unprecedented work has led to an understanding of the importance of genes as inherited material, as the molecule that stores the information from the history of life, and as determinants of the traits of organisms. Before genes were discovered and understood, it was difficult to explain inheritance and the evolution of organized traits. Genetics has become the central, theoretical organizing principle of biology.
However, recent work, and ideas that will be considered in this book, raise tempering questions about several aspects of the present view of genes. First, the connection between genes and traits is in many ways more indirect and subtle than most biologists have thought (or than many still seem to think). Second, although DNA is one of the most important and widespread constituents of new organisms, most of which begin life as single cells, some aspects of inheritance are not strictly based on DNA sequence. These include parental RNA and proteins, DNA packaging and modification, and other chemical characteristics of the cell. Many aspects of behavior and the construction of environments are also inherited (Chapter 3), and additional nongenetic aspects of inheritance may be identified in the future. Despite the undeniable and continued power of genetics to organize biological thinking and research and to account for the evolution of life, the degree to which variation, inheritance, or evolution should be described strictly in genetic terms is a more open question.
Both Darwin and Gregor Mendel, the Moravian monk who first characterized the segregation of heritable traits with his experiments on pea plants, worked with very deterministic notions of "genes" (the term itself was not yet in use, of course), that is, with the heritable determinants of phenotypes. Darwin's theory of pangenesis turned out to be quite wrong, perhaps because he was thinking of a gradual change of continuous variation. Mendel, however, chose traits with simple transmission patterns that bred true in hybrids and that were essentially controlled by single "factors." That knowledge eventually led us to the genes themselves.
Mendel's discovery transformed biology, but in some ways we have become entrapped by the elegant simplicity of his choice of dichotomous traits manifesting "mendelian" inheritance. Many of the driving concepts of biology are built on this, and we still often pay little more than lip service to what goes on between the genotype an organism is born with and the phenotypes that develop in its lifetime. This is especially true of phenotypes not closely connected to the protein products of individual genes—the ones of most interest to Darwin and evolutionary biologists ever since.
In fact, the inner workings of life are far more complex than had been expected. In important ways, we attempt to force classical darwinian-mendelian theory in circumstances in which the fit is not so good. This does not mean that inherited traits do not involve genes, but genes are not always good predictors of traits, as will be seen in Chapters 3 through 5. Put another way, the mendelian inheritance of genes does not imply the mendelian inheritance of phenotypes (traits). Phenotypes are not inherited; organisms begin life as single cells with genes but not with arms, stomachs, or flowers.
Determinism, Reductionism, and Genetics
One of the fundamental aspects of most Western science is reductionism. The idea, attributed generally to influential thinkers such as Descartes and Francis Bacon, derives from our notions of empirical experimental design that the phenomena of nature can—indeed, perhaps should—be studied and understood part by part, ultimately all the way down to the most fundamental parts. This does not mean that each part acts independently, nor that we necessarily ever will understand all aspects of a trait. But it does assume that in principle we can come to a fundamental understanding of a phenomenon by isolating and analyzing its component effects. Just the way we disassemble a machine into its parts (discussed earlier).
The ultimate belief of reductionism is that the universe is (only) a space filled with matter and energy. If this view is true, then everything can in principle be "reduced" to, that is, ultimately explained in terms of (only), molecular and energetic phenomena. Biological phenomena, too, will ultimately be understood best in terms of the molecular biology of genes, the "atoms" of biological information (in some ways, biochemists would extend this even further down, of course). In this spirit, geneticists seek to study each trait in terms of genes, as separable causal elements. The objective is to explain a phenotype in terms of the effects of the individual genes that affect it—just the way we reassemble a machine from its parts.
Reductionism in biology works at various levels. Functional anatomists attempt to reduce traits like locomotion to the contribution of individual bones and muscles. To a psychologist, learning may be comprised of recognition, memory storage, memory recall, units of meaning, and ultimately to neurons and neurotransmitters. To an ecologist, an ecosystem consists of predators, prey, a food chain, etc.
Each biologist chooses his/her level of reduction. Some may choose to look only at the role of frogs in the biodiversity of a swamp without feeling compelled to try to explain the croaking of the frog in terms of its genes. Some wish to go farther and to "reduce" the croaking to hormone receptors, neuronal pathways, and the like. A biochemist may see this all as a problem in ligand-receptor binding, signal trans-duction, and gene regulation by action potentials of auditory hair cells.
Some biologists, although acknowledging that one can account for a frog in terms of chemicals, believe that reductionism cannot adequately explain the croaking. From this point of view, the phenomenon is an emergent one, that must be understood at its own level of organization. Field biologists may not even care to try to understand a bullfrog's croak and his mate's response in terms of DNA sequence data or hormone kinetics, a level of accounting in which they have no interest—any more than you might think the words you are reading could be understood by analyzing the chemistry of their ink and paper.
To some reductionists, higher-level studies barely count as important science, as they are too superficial. A common reason given is that we are not as good at making "operational" the study of complexity as we are at reductionist, experimental methods. The latter have a long history, and the triumphs of modern science and technology are the fruit. This view holds that higher-order phenomena are not fundamental and that eventually we will be able to predict "emergent" biological—or even cultural—traits by analyzing their components (e.g.,Wilson 1998). Science does not allow nonmaterial causation, so how can an understanding of any phenomenon of nature not follow from an adequate understanding of its parts?
A reductionist perspective does not assert that causation is always one-to-one, but only that if we know all the actors, we will explain the play. An illustration of the issues involves gene action. Geneticists have long recognized that genes are often pleiotropic, that is, have many functions. Similarly, different genotypes can be found in individuals with the same phenotype. The genotype-to-phenotype relationship is often many-to-many in nature. Even if each component can be characterized in molecular terms, the overall effect of a gene on an organism, or of natural selection on a gene, may depend on the set of interacting constraints. We may not be able to predict the trait from any one of its components, but we should be able to do so from the set. An important question is when or how well we can ascertain or even define what that set is.
Arguments about reductionism are not new, and they are probably not resolvable, but the points are important in this book, whose aim is to understand the role of genes in how organisms manage their lives. But what it means to "understand" a phenomenon depends to a great extent on the question being asked.
Conservation, variation, and homology
With caveats raised earlier about horizontal transfer, evolution is generally diversifying over time. Based on this, a central organizing fact in evolutionary biology, indeed a key to what Darwin and Wallace contributed, is homology, the conservation of traits in descendants of a common ancestor. It is historical connectedness that differentiates evolution from creation. In the 20th century, as genes took the throne of the biosphere, the key element of homology tended to shift from shared traits to shared genes or DNA sequence elements. However, the more we learn about genes and how they are used, the more rather than the less elusive the concept of homology has become (e.g., Hall 1999;Wagner 1999; Wagner and Gauthier 1999).
For example, the limbs of tetrapods are considered to be descendants of fins in the common fishlike ancestor of these four-legged vertebrates, so limbs and fins are homologous in the classical sense. As expected, homologous genes are involved in limb development in different tetrapods. By contrast, eyes appear to have evolved independently in insects and vertebrates and have long been used as an example of analogous traits. However, developmental geneticists have recently discovered unexpected similarities in the mechanisms used to generate eyes in both groups of species—but only in some of the mechanisms and not all the same ones in all species. We will discuss this specific question in detail (e.g., Chapter 14). Here, the point is that recent discoveries of the genetic connectedness of diverse life forms show the need for a reformation of the very important homology-analogy question.
Ever since Mendel, and reinforced by the Central Dogma that one gene codes for one protein, the view that genes evolve "for" traits has been predominant. But this can be misleading in at least two important ways. On the one hand, unlike the eye situation, the genetic basis of similar, seemingly homologous traits sometimes turns out to be different (e.g., Hall 1999; Raff 1999; Wagner and Misof 1993; Weiss 2002; Weiss and Fullerton 2000). When this happens, the trait itself may be homologous in the traditional sense but not its underlying genes. On the other hand, it may be that a trait is produced by a developmental process that is completely conserved (homologous among species under comparison) but that the details of the process vary. For some traits with multiple elements, like body segments, the process and the nature of the trait may be conserved but the individual elements may not be homologous.
Another common view is that traits that have been shared since lineages diverged from a common ancestor have been conserved by natural selection. However, it is possible that the trait is conserved simply by the genealogical relationship. Even if selection is not acting, a trait will only change slowly between related lineages. Genes of humans and mice are around 80 percent identical on average, even in regions of the genome unlikely to have been seriously affected by natural selection. Perhaps of more potential importance, there are reasons to believe that in some instances, involving either particular chemical interactions or the interactions of many components, natural stable states or "attractors" may exist (such as "positions of organic stability" mentioned above). These may conserve a trait over long time periods or even provide an element of inevitability (e.g., Kirschner et al. 2000; Laughlin et al. 2000; Monod 1971; Morowitz et al. 2000; Schuster 2000)—in fact, as noted earlier, many aspects of life that seem complex may not be all that improbable in the first place (Keefe and Szostak 2001; Schuster 2000). Furthermore, a phe-notype may be conserved but not its underlying genetic basis; the molecular structure of transfer RNA may be one example (Fontana and Schuster 1998). Similarly, as referred to earlier, natural chemical constraints would not need competitive natural selection in the usual sense, to be maintained. Indeed, once reached, it may be very difficult to get out of such canalizing constraints to try another way.
On Being a "Being"
For many centuries, the array of natural organisms was viewed in Western thought as a natural hierarchy, often referred to as the Great Chain of Being. The longstanding view was that a creation event had produced this natural order of life. In this book, we will see how the same sense of relationships was transformed by the idea of evolution. However, we will suggest ways in which, even from a genetic and evolutionary perspective, the notion of "being" has been too restrictive.
Biology is about beings, plural, not just the more philosophical phenomenon of
"being"—the existence of life forms. The term is generally applied to organisms in the colloquial as well as scientific sense: a bird, ant, person, or birch tree. However, in many ways this is arbitrary: species, cells, and even genes can also be viewed as "beings." At these various levels of observation, the interactions within and among these entities are logically similar and, as we will see in various chapters, involve similar or identical genetic mechanisms. A broader, more flexible sense of the term "being" helps unify biology and make the origin of its complexity more readily understood.
WHY THESE CONCEPTUAL ISSUES ARE IMPORTANT This chapter raised a number of issues, caveats, and perspectives that may impact our ability to understand the fundamental generalities of complex life. It is important to examine even the most basic of what are often tacit assumptions. Exceptions and alternative viewpoints can have considerable if sometimes unexpected merit, as we will try to show in various ways throughout the book.
The purpose of searching for generalizations about life is to be able to extrapolate from observations necessarily limited to only a subset of individuals, species, or model systems to as broad a scope as possible. Without observing everything, how far can we extrapolate? The answer is important for animal model work, agriculture, and biomedicine, as well as for understanding basic biology.
We would also like to reconstruct the unique history of life by taking specific account of the trace of past events left in DNA sequences or in their indirect manifestation in biological traits. For more than a century, this has been based on a strong branching model, but suppose this is not the right model? Suppose homology loses its clear-cut meaning before we get too far in the past, or life turns out not to have a single-trunked genealogy, or that genes turn out to be less determinative than has been thought.
In a historical field like evolutionary biology, the challenge of making retrospective evolutionary reconstructions is daunting. Unlike chemistry or physics, almost any general statement will have nontrivial exceptions, and for many biological observations there are multiple comparably plausible explanations (for example, chance, competition, and cooperation), and it is all the more important to put our notions to the test.
Classic darwinian theory has been exceedingly powerful at providing coherent explanations and has transformed thinking in biology essentially by equating similarity with genealogy. We caution against the overuse of the old Cartesian notion that an organism is a machine, but with linguistic irony our own evolutionary explanations are fabricated, in the literal sense. We make them up, hoping they are accurate re-constructions of a species' or trait's history. But we can rarely be too sure.
The historical connectedness of organisms and the consequent storage of historical information in genes provide unifying tools for understanding. But we can still ask important questions about how apparent order can emerge from the disorder of an undirected universe.
Was this article helpful?